LCS Orchestrator Read Caching for Exclusive Storage Access
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional information handling systems face latency issues during data storage operations due to supplemental storage services like decompression and decryption, and caching data across multiple host paths requires complex coordination and incurs high overhead.
Innovation Solution
An Information Handling System (IHS) with a processing system and memory that includes instructions to identify exclusive access to a storage subsystem, activate read data caching for a client device, retrieve data from the cache, and deactivate caching when access is no longer exclusive, thereby optimizing data provisioning.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of time
If read data caching is activated for storage subsystems accessible by multiple client devices, then read latency is reduced, but cache coherency issues and storage operation overhead increase
Solution Approach 1:
The patent applies local quality by enabling caching only for storage subsystems that are exclusively accessible by a single client device, while disabling caching for storage subsystems shared by multiple client devices. This selective approach optimizes read performance for exclusive-access scenarios without introducing cache coherency complexity in multi-client scenarios.
2Loss of time
If read data caching is activated for storage subsystems, then read latency is reduced, but storage operation overhead increases
Solution Approach 1:
The patent implements local quality by dynamically adjusting caching behavior based on the access pattern of each storage subsystem. For storage subsystems with exclusive single-client access, caching is enabled to reduce read latency. For storage subsystems with multi-client access, caching is disabled to avoid the overhead of cache coordination algorithms, thus maintaining storage operation throughput.
3Speed
If caching is enabled for storage subsystems with multiple host paths, then read performance improves, but cache coherency maintenance becomes complex
Solution Approach 1:
The patent applies local quality by differentiating between storage subsystems based on their access characteristics. For storage subsystems accessible by multiple host paths, the patent disables caching to avoid the complexity of cache coordination algorithms. This selective disablement maintains simplicity while preserving read performance for multi-access scenarios where caching would be problematic.
Data Source
AI summary
A Logically Composed System (LCS) data provisioning system includes an orchestrator device that includes a cache subsystem and that is coupled to client devices and storage subsystem(s). When the orchestrator device identifies that a first client device has exclusive access to the storage subsystem(s), it activates read data caching for the storage subsystem(s). The orchestrator device then receives a first read request from the first client device that is directed to first data that is stored in the storage subsystem(s) and, in response, retrieves the first data from the cache subsystem and provides the first data to the first client device. When the orchestrator device identifies that the first client device no longer has exclusive access to the storage subsystem(s), it deactivates the read data caching for the storage subsystem(s).